# Claude Opus 4.5 vs GPT-4o

> Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 28.5 on the Noometry Index. GPT-4o costs 2.3× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-5-vs-gpt-4o
- Last updated: 2026-10-10
- Shared benchmarks: 44

## Summary

- They share 44 benchmarks with published results for both. Claude Opus 4.5 scores higher in 9 categories and GPT-4o in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.5 leads 42.5 to 9.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.1% for Claude Opus 4.5 and 6.4% for GPT-4o.
- GPT-4o is cheaper at $2.50 / $10 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
- Claude Opus 4.5 accepts more context: 200K tokens versus 128K.

## Snapshot

| | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.5 | 28.5 |
| Rank | 39 | 316 |
| Context | 200K | 128K |
| Input $/M | $5 | $2.50 |
| Output $/M | $25 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4.5: 55.4 (#21)
- GPT-4o: 25.1 (#323)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| SWE-bench Verified | 76.7% | 31% |
| SWE-bench Verified (bash only) | 76.8% | 21.6% |
| GSO | 26.5% | 0% |
| WeirdML | 63.7% | 25.1% |
| LMArena Coding | 1504 | 1297 |
| Aider Polyglot | — | 45.3% |
| LMArena WebDev | 1493 | — |
| SWE-bench Multilingual | 70.7% | — |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| ALE-Bench | 1,025 | — |
| AlgoTune | 1.77 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |

## Agentic & Tool Use

- Claude Opus 4.5: 47.2 (#11)
- GPT-4o: 20.8 (#133)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| GDPval | 45.5% | 9.9% |
| Cybench | 82% | 12.5% |
| BALROG | 43.5% | 32.3% |
| LMArena Search | 1180 | 1006 |
| METR Time Horizons | 75% | 40.8% |
| Terminal-Bench | 63.1% | — |
| Berkeley Function Calling Leaderboard | 77.5% | — |
| Remote Labor Index | 3.8% | — |
| TheAgentCompany | — | 8.6% |
| τ²-bench Airline | 84% | — |
| τ²-bench Banking | 24.7% | — |
| τ²-bench Retail | 79.6% | — |
| τ²-bench Telecom | 92.3% | — |
| DeepResearch Bench | 54.8% | — |
| OSWorld | 66.3% | — |
| Vending-Bench 2 | 4,967 | — |

## Reasoning

- Claude Opus 4.5: 42.5 (#43)
- GPT-4o: 9.4 (#338)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 37.6% | 0% |
| SimpleBench | 62% | 17.8% |
| ARC-AGI-1 | 80% | 4.5% |
| Chess Puzzles | 12% | 13% |
| EnigmaEval | 11.9% | 0.8% |
| LMArena Hard Prompts | 1476 | 1281 |
| DTBench | 89.9% | 64.5% |
| LMCA | 44.5% | 16.6% |
| Epoch Capabilities Index | 150.1 | 128.97 |
| ForecastBench | 60.7 | 57.7 |
| Kagi LLM Benchmark | 80.2% | — |
| NYT Connections (extended) | 52.5% | — |
| CritPt | — | 0% |
| EBR-Bench | 14.3% | — |
| LiveBench Reasoning | — | 55.8% |
| Mystery Game Puzzles | 22% | — |
| LiveBench Data Analysis | — | 60.9% |
| LiveBench | — | 55.3% |

## Math

- Claude Opus 4.5: 38.5 (#85)
- GPT-4o: 10.2 (#303)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| FrontierMath (Tiers 1-3) | 34.4% | 0.4% |
| OTIS Mock AIME 2024-2025 | 86.1% | 6.4% |
| LMArena Math | 1463 | 1285 |
| FrontierMath (Feb 2025 set) | 20.7% | 0.3% |
| FrontierMath Tier 4 | 4.9% | — |
| ProofBench | 36% | — |
| Omni-MATH | — | 29.3% |
| LiveBench Math | — | 49.5% |
| MATH Level 5 | — | 53.3% |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Opus 4.5: 56.4 (#38)
- GPT-4o: 28.8 (#235)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| GPQA Diamond | 86% | 49.2% |
| Humanity's Last Exam | 25.2% | 2.7% |
| SimpleQA Verified | 45.7% | 26% |
| Vectara Hallucination Rate | 10.9% | 9.6% |
| LMArena Expert | 1488 | 1250 |
| MMLU-Pro | — | 71.3% |
| Confabulations | — | 15.3% |
| GPQA (HELM) | — | 52% |
| MMLU | — | 88.1% |

## Multimodal

- Claude Opus 4.5: 31.1 (#99)
- GPT-4o: 33.9 (#85)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| GeoBench | 75% | 71% |
| VPCT | 40% | 40% |
| LMArena Vision | — | 1137 |
| Video-MME | — | 71.9% |
| Furniture Assembly | 28.3% | — |
| LMArena Document | 1462 | — |
| ScienceQA | — | 88.5% |

## Multilingual

- Claude Opus 4.5: 54.5 (#45)
- GPT-4o: 43.4 (#181)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| LMArena Non-English | 1438 | 1283 |
| LMArena Chinese | 1470 | 1277 |
| LMArena French | 1471 | 1303 |
| LMArena German | 1449 | 1282 |
| LMArena Japanese | 1415 | 1257 |
| LMArena Korean | 1424 | 1234 |
| LMArena Russian | 1448 | 1286 |
| LMArena Spanish | 1455 | 1292 |

## Instruction Following

- Claude Opus 4.5: 77.7 (#20)
- GPT-4o: 66.8 (#202)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| LMArena Instruction Following | 1478 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
| IFEval | — | 81.7% |

## Long Context

- Claude Opus 4.5: 46.4 (#21)
- GPT-4o: 39.4 (#174)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1480 | 1289 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | 21.1% | — |

## Writing & Preference

- Claude Opus 4.5: 68.2 (#28)
- GPT-4o: 52.8 (#160)

| Benchmark | Claude Opus 4.5 | GPT-4o |
|---|---|---|
| LMArena Text | 1451 | 1300 |
| LMArena Creative Writing | 1445 | 1291 |
| LMArena Multi-Turn | 1466 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| EQ-Bench Creative Writing | 1687 | — |
| WildBench | — | 82.8% |
| LiveBench Language | — | 47.6% |

## FAQ

### Is Claude Opus 4.5 better than GPT-4o?

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 28.5 on the Noometry Index. GPT-4o costs 2.3× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4.5 or GPT-4o?

GPT-4o is cheaper. It lists at $2.50 per million input tokens and $10 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

### Is Claude Opus 4.5 or GPT-4o better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 55.4 versus 25.1 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 4.5 does, with 200K tokens against 128K.

### How many benchmarks do Claude Opus 4.5 and GPT-4o share?

44 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and GPT-4o has 72.